Some Considerations on the Use of Animals in Vaccine
Production and Quality Control
Coenraad F.M. Hendriksen
Introduction
Over the past decade, debate over the use of laboratory animals has steadily increased, both in public circles and within the scientific community. While proponents of animal experimentation argue that animals have instrumental value, many also acknowledge that animals have intrinsic value. This has resulted in an ethical re-appraisal of animal experiments. Currently, it is generally accepted that the moral admissibility of animal experimentation should be judged by weighing that experiment's anticipated benefits for humans against the anticipated discomfort in the animals,(1,2) One field of research criticized for its extensive use of laboratory animals is vaccine production and quality control. Some people question whether the practical benefits of immunization outweigh the moral costs in laboratory animal lives.(3) This paper presents data to help assess the (public health) benefits vs. (moral) costs. It discusses the rationales for using animals in vaccine production and quality control, evaluates the effects of immunization on public health, and addresses society's demand for safe and reliable products. It also recognizes the inherent moral dilemma of using laboratory animals. Finally, it reviews some opportunities for replacement, reduction, and refinement.
Historical Background
Historically, there has been a close and longstanding association between microbiology and vaccine development and the use of laboratory animals. Robert Koch's postulates, published in 1884 in his study "The Etiology of Tuberculosis" have been fundamentally important. These postulates describe the criteria used to establish the causal relation between a microorganism and a disease. One of the criteria is that inoculation of pure culture of the suspected microbe in a healthy, susceptible animal leads to characteristic symptoms of the disease. Koch's postulates gained general acceptance in microbiology and helped to undermine the theory of spontaneous generation, which had long been used to explain the origins of diseases. Robert Koch (1843-1910) and, Louis Pasteur (1822-1895) independently succeeded in scientifically demonstrating the causal relation between infectious diseases and microorganisms. Due to the discoveries of Pasteur, Koch, and other scientists of the mid-nineteenth century, the causes of many communicable diseases were established by the turn of the century. These insights into etiology, gained in part through animal experimentation, opened the way to a rational approach toward fighting the most threatening diseases. Previously, therapies had largely been based on tradition or empiricism and had seldom been effective. Tetanus, for instance, was said to be cured by bloodletting, giving quinine and wine in large doses, or the use of a cold bath.(4) In fact there was no cure or immunization for tetanus or any other infectious disease (except Jenner's smallpox vaccine) until the end of the nineteenth century.
Pasteur's vaccine against fowl cholera, discovered in 1880, was a major breakthrough in prevention and therapy. Pasteur first injected chickens with old cultures of Pasteurella multocida, the causative agent; very few clinical symptoms developed. A subsequent challenge with virulent microorganisms produced no adverse effects. The vaccine against fowl cholera thus became the first experimentally developed vaccine. The scientific community was at first skeptical about the benefits of vaccination; few recognized its importance. In 1881 the American scientist Salmon wrote:
Who can estimate the value to the world in human lives and property of a method by which each virus may be transformed into a vaccine and made to protect people as well as animals from destructive plagues which so frequently cause death and destruction over vast sections of the world?(5)
Following Pasteur's pioneering work, several researchers initiated and extended the use of immunization as a means of combatting infectious diseases. In most cases, experiments on animals played an essential role: first, in developing models by which to understand the (immuno) pathogenesis of the infectious disease; next, in selecting of vaccine strains and testing for immunogenic properties; finally, in improving the quality and control of the immunologicals. For example, the German scientist Behring (1854-1917) used large numbers of guinea pigs, mice, rats, dogs, and other animals for his work on diphtheria antiserum. Because animal experimentation played an important role in the development of the diphtheria antiserum public acceptance of animal research subsequently increased.(6) Many of today's widely practiced, routine animal testing procedures, such as potency testing and the use of a standard preparation, originated in diphtheria vaccine development and testing.
Today, many immunologicals are available for preventing and treating human and animal infectious diseases (Figure 1). Unfortunately, large numbers of laboratory animals are required for their development, routine production, and quality control.
Figure 1: Human vaccine development [dates approximate]
Extent and Purpose of Animal Use in Vaccine Production and Quality Control.
According to the 1989 statistics on the use of laboratory animals in The Netherlands, 207,147 animals were involved in the production, testing, and biological standardization of sera, vaccines, and other biological products. This represents about 20% of the total number of animals (1,010,919) used that year.(7) The statistics do not specify the category of biological products, but the number of animals used for vaccines alone is estimated to be at least 125,000.(8) No data are available from other countries; however, taking into consideration the universal importance of vaccines, this use of animals is probably at comparable levels elsewhere. The extensive use of animals is related to the biological origin of most vaccines. Because of risks of contaminants, impure products, and production failures, every batch is tested for safety and potency. In order to regulate vaccine quality control, national regulatory agencies, such as the Food and Drug Administration, as well as international institutions, such as the World Health Organization, have established guidelines.
Table 1 addresses a different issue -- the degree of distress inflicted on animals for purposes of production and quality control of biologicals as compared to other categories of animal experimentation in The Netherlands. As can be seen, the biologicals category involves severe distress to the largest number of animals.
Table 1. Numbers of experimental animals used in 1989 in The Netherlands for biologicals and other fields of research, categorized by distress rating (7)
| Field of research | Distress |
||
Slight |
Moderate |
Severe |
|
| Number % | Number % | Number % | |
| Biologicals | 74,236 (36) | 41,056 (20) | 91,855 (44) |
| Toxicology | 37,159 (37) | 17,528 (18) | 43,859 (45) |
| Pharmacology | 151,201 (67) | 43,591 (19) | 29,917 (14) |
| Diagnosis | 5,797 (22) | 2,793 (11) | 17,391 (67) |
| Education | 8,143 (69) | 3,117 (26) | 552 (06) |
| Basic Research | 241,989 (55) | 128,520 (29) | 72,215 (16) |
| Total | 518,525 (52) | 236,605 (22) | 255,789 (25) |
Most animals used in vaccine research are used for quality assurance purposes. In vaccine production, animals play only a minor role because they are needed only to provide tissue for cell cultures. The number of animals used for production is declining because vaccine-virus propagation in primary cell cultures is gradually being replaced by propagation in subcultured cells or cell lines.
In the area of quality control, animals are generally used in two kinds of tests -- safety tests and potency (efficacy) tests. Safety testing is designed to protect vaccine recipients from extraneous agents, toxicity, and (in the case of live attenuated vaccines) from residual virulence. Toxicity can be specific (related to the vaccine agent) or abnormal (related to unknown contaminants in the vaccine product).
The aim of potency testing is to ensure that the vaccine induces a protective immunity. The potency of live vaccines such as BCG or oral polio is determined by germcount or virus titration and is performed totally in vitro. However, the potency assay of the inactivated vaccines, such as the major human bacterial vaccines, is based on in vivo procedures. Most animals used in vaccine production are for potency testing, and these tests cause severe distress, as can be seen in Table 2. Clearly, large-scale immunization has involved enormous animal suffering, in terms of both absolute numbers and degree of distress inflicted.
Table 2. The number of animals required and distress rating for various tests in the quality control of diphtheria, tetanus, and pertussis vaccines in The netherlands in 1985.
| Animal test | Approximate number of animals |
Distress rating |
| Potency* | 46,000 |
severe |
| Specific toxicity | 1,200 |
slight to severe |
| Abnormal toxicity | 250 |
slight |
*Data on the potency test are based on the lethal challenge procedure.
If one accepts the view that the moral admissibility of animal experimentation should be based on a cost-benefits analysis as well as the view that animal use is crucial to vaccine production and quality control, one must question whether the benefits of immunization to humans outweigh costs to animals. Some critics have raised doubts about the value of vaccination.
Skepticism Regarding Immunization's Public Health Benefits
Since the development of the first vaccines, some people have opposed vaccination. In late nineteenth-century Victorian society, an active antivaccinationist movement attacked such immunization research as Pasteur's work on rabies and Behring's development of diphtheria antiserum. Ideologically, the antivaccination movement was closely allied to the sanitary movement, which urged hygienic measures to improve public health. Some of the anti-vaccinationists were also prominent anti-vivisectionists, with a strong aversion to the animal experimentation used in vaccine development.(9)
Arguments against immunization have not changed much during the last century. In his book Medical Nemesis, Ivan Illich (10) disputes the role of drugs and immunologicals in the overall improvement of public health. In a well documented study, McKeown (11) arrives at the same conclusion, stating that immunization and therapy hardly had any impact on the death rate due to infectious diseases, with the exception of diphtheria antiserum. His conclusion is based on statistics showing that mortality due to common infectious diseases (tuberculosis, measles, whooping cough, and scarlet fever) had fallen to a relatively low level before immunization was employed. McKinlay and McKinlay have analyzed the effect of medical measures in combatting influenza, pneumonia, diphtheria, whooping cough, and poliomyelitis; they estimated that, at most, 3.5% of the total decline in mortality was due to vaccination against those diseases since 1900.(12) Similarly, Sharpe (3) has concluded that vaccination has reduced the death rate only marginally.
The Effects of Immunization on Public Health
Undeniably, mortality rates were already declining before the introduction of immune therapy and prophylaxis at the end of the nineteenth century. (Table 3 shows mean annual mortality figures of two cities during successive historical time periods.) This increased life expectancy, due primarily to a reduction in deaths from infectious disease, is chiefly attributed to improvements in nutrition, living and working
Table 3. Annual mortality per 1000 inhabitants in London, England and Göteborg, Sweden during successive time periods (13)
London |
Göteborg |
||
| Time interval | Mean annual mortality* |
Time period | Mean annual mortality* |
| 1620-1643 | 70 |
1776-1800 | 44 |
| 1728-1757 | 52 |
1801-1825 | 38 |
| 1800-1810 | 29 |
1851-1875 | 28 |
| 1871-1880 | 19 |
1876-1885 | 20 |
*Deaths per 1000 inhabitants
conditions, hygiene, and sanitation.(11) For example, death rates from whooping cough and measles fell continuously throughout this century, and mortality was already at a very low level before large-scale immunization began (Figures 2,3). Given this data, has animal experimentation (insofar as it has contributed to vaccination) had any impact on the decline of mortality from measles and whooping cough? Also, should the use of laboratory animals for quality control of measles and whooping cough vaccines be reconsidered if immunization no longer leads to lower mortality rates?
The value of improved sanitation and hygiene in decreasing mortality rates from measles and whooping cough is beyond dispute. However, these improvements derived in part from animal experimentation, performed by bacteriologists and immunologists. Scientific proof of the "germ-theory," using Koch's postulates, led to new insights into disease transmission and pathogenesis; these insights led, in turn, to improved hygiene.
Because death rates from measles and whooping cough were already at low levels when immunization was introduced, it is inappropriate to assess the effects of the vaccines in terms of decreased mortality. It is more meaningful to examine morbidity rates and incidences of adverse reactions. Data are available for both diseases. According to Banatvala, 90,000-100,000 cases of measles are reported each year in the UK., and 10% of patients experience complications.(14) The protection given by measles vaccination is documented in Table 4. Measles vaccination in the USA is estimated to have a benefit-cost ratio of 10:1; prevention through vaccination is estimated to save approximately $500 million annually.(15)
Table 4. Benefits of measles vaccination
Incidence (Number per 100,000 cases following: |
|||
Adverse reaction |
Infection |
Vaccination |
Ratio, infection/ vaccination |
| Pneumonia | 5000 |
1.000 |
5000 |
| Convulsions | 720 |
60.000 |
12 |
| Brain damage | 280 |
0.100 |
2800 |
| Early death | 10 |
0.020 |
500 |
| Chronic infection, delayed death |
2 |
0.075 |
26 |
Like measles immunization, pertussis immunization can be evaluated in terms of benefit versus risk. In the U.S., the incidence of pertussis began to decline prior to widespread use of pertussis vaccine, but the rate of decline was greater after routine immunization was introduced.(17) If the mortality trends from 1900 on had persisted -- with 16% average declines for infants in each five-year period and with 21% declines for children one to four years old -- 8,000 deaths from pertussis in U.S. children under five years of age would have been expected between 1970 and 1974 without vaccination. In fact, only 52 were reported.(18) The U.K controversy in 1974/1975 about pertussis vaccination illustrates immunization's benefits. Due to widespread discussion of risk vs. benefit in both public and scientific sectors, public acceptance of pertussis vaccine decreased dramatically. In 1974 the vaccination rate was 75%, but by 1978 it was only 30%(17) A rise in pertussis notifications accompanied this decrease in vaccine acceptance, with notifications increasing by a factor of ten to sixteen.(19) Based on these figures, the U.K. DHSS Joint Committee on Vaccination and Immunization concluded in 1981 that the risks of serious complications to immunization were slight and were outweighed by its advantages.(20) (Table 5 provides data on risks vs. benefits in pertussis vaccination.) Based on data in prosperous countries, then, this analysis leads to the conclusion that vaccination is beneficial
Table 5. Benefits of vaccination against natural whooping cough.
| Incidence (Number per 100,000 cases) following |
|||
| Adverse reaction | Infection | Vaccination | Ratio Infection/vaccination |
| Shock | -- | 15.0 | -- |
| Convulsions | 4,000 | 45.0 | 89 |
| Encephalitis | 2,000 | 1.5 | 1,300 |
| Permanent brain damage | 1,300 | 0.3 | 4,300 |
| Death | 2,000 | 0.2 | 10,000 |
| Data taken from Galazaka et al. (16) | |||
Infectious Diseases and Immunization in the Developing World
Many vaccine-preventable diseases now rare in technologically advanced nations are still common in the developing world. Neonatal tetanus, for instance, persists in developing countries; in some areas it accounts for 20%-70% of neonatal mortality.(21) Estimates of annual mortality due to a number of infectious diseases are given in Table 6.
Table 6. Estimated annual morality due to vaccine-preventable diseases in developing countries.
| Disease | Number of deaths | |
| Measles | 2,000,000 | |
| Tetanus | 800,000 | |
| Pertussis | 600,000 | |
| Tuberculosis | 30,000 | |
| Diphtheria | 25,000 | |
| Poliomyelitis | 250,000* | |
| *Cases of paralysis, not deaths. Adapted from Poore (22) | ||
The high incidence of mortality from vaccine-preventable diseases is ascribed to a combination of infection, malnutrition, poor sanitation, and inadequate housing. Children often have their defense mechanisms compromised by low birth-weight and then by a series of stresses, including whooping cough, weaning, and repeated episodes of diarrhea and malaria.(8) Improvements in housing, nutrition, and sanitation would certainly go far in combatting these diseases. Until these improvements are achieved, however, vaccines can be of substantial benefit. Vaccines are simple to administer and highly effective, They can also be combined with other services needed by children, such as growth monitoring, treatment of diarrhea, nutritional advice, and information on sanitation.(8) In 1977, to address the problem of mortality and morbidity in early childhood, the WHO created the Expanded Programme on Immunization (EPI), with the ultimate goal of making immunization services available for all children of the world. Before 1974, probably less than 5% of infants were protected by vaccines; today at least 50% of the infants receive a combined diphtheria/pertussis/tetanus vaccine as well as a polio vaccine. It has been estimated that immunization annually prevents more than a million deaths from measles, pertussis, and neonatal tetanus.(22) EPI's impact on worldwide laboratory animal use is unclear because no precise data are available. Abdussallam (23) has estimated that EPI has increased the number of animals used by a factor of two to five.
In summary, immunization seems an important part of effective public health care, in both the industrialized and developing countries. In industrialized countries, immunization against measles and whooping cough has reduced the number of notifications and the incidence of adverse reactions, even if it has not contributed to a decline in death rate. In developing countries, immunization is a prerequisite for reducing infant mortality.
Given our present knowledge of vaccine technology and the importance of immunization, the use of laboratory animals in vaccine development and quality control is unlikely to cease in the near future.
Public Pressure, Vaccine Development, and Quality Assurance
Society's demand for safe and reliable vaccines has contributed to the current extensive use of laboratory animals. Frequenfly, quality assurance tests are modified or introduced after accidents arouse public attention. For example, in the "Cutter-case" (named after an American manufacturer of polio vaccine), 250 people contracted poliomyelitis from an incompletely inactivated polio vaccine; five died.(24) This tragedy resulted in stricter guidelines regarding testing for residual live virus. The intracerebral monkey test was extended to include an intraspinal and intramuscular injection.
A second incident involved adverse reactions to pertussis vaccination in the U.K. in 1974. Suddenly, the public feared immunization's potential harm to children. The media highlighted controversy among experts,(25) and immunization rates fell from 75% in 1974 to below 40% in 1976.(17) Subsequently, whooping cough rose sharply. Research efforts to develop a nontoxic pertussis vaccine also received highest priority. Laboratory animals used in this effort number in the millions.
Even if we conclude that the benefits of animal research to humans outweigh animal suffering and death involved, we are not discharged from our moral responsibility to perform research that maintains the highest level of animal care. Above all, we should make every effort to apply the "3R" concept to animal experimentation -- replacement, reduction, and refinement.
Opportunities for Replacement, Reduction, and Refinement
Elsewhere, I have reviewed the opportunities for replacement, reduction, and refinement of animal experimentation in the area of vaccine production and control.(26) When vaccines are produced according to Good Manufacturing Practice (GMP) requirements, tests for abnormal toxicity are unnecessary and should be omitted from the protocol. These tests are intended to ensure that no (chemical) contamination has occurred during the vaccine filling process. Each batch of vaccine is tested by inoculating two guinea pigs and five mice with a sample of the final product, after which the animals are observed for seven days for weight loss and signs of toxicity. Lack of specificity, however, frustrates the interpretation of results. Consequently, the scientific value of such tests is questionable.
Another test that should be critically re-evaluated is the potency test on diphtheria and tetanus toxoid vaccines. In most countries (an exception being the USA), this test is based on a quantitative approach, involving three doses of a test preparation and three doses of a reference preparation. This so-called "3 + 3" type of assay is a statistically valid test giving reliable and precise results. However, precise results may not be needed, since WHO and European Pharmacopoeia requirements state only that potency estimates should exceed a minimal level of international units. Present-day toxoid vaccines are potent, and defined products and consistency in production are well-established in most centers of production. Therefore, a simplified potency test for routine control of toxoid batches might be considered. Such a test could be based, for instance, on a "1 + 1" assay, immunizing one group of animals with one dose of the test vaccine and one group of animals with one dose of the reference preparation. Combined with replacement of the lethal potency test that involves use of a serological approach, the one dose (serological) test could significantly reduce both number of animals used and the degree of distress.
Some further opportunities for reduction and/or refinement of animal experimentation in potency testing of vaccines are summarized in Table 7. Other opportunities for replacing, reducing, or refining animal use include standardization of international testing guidelines, re-evaluation of test procedures, development of new vaccine production technologies, and wider use of in vitro models. Some of these possibilities, however, need further elaboration or validation before they can be introduced in routine quality control.
Table 7. Alternatives in potency testing of human bacterial and viral vaccines.
| Vaccine | Alternative method | Reference |
| Diphtheria | Vero cells Toxin Binding Inhibition |
Kreeftenberg et al. (27) Hendriksen et al. (28) |
| Tetanus | Lf test Toxin Binding Inhibition |
Lying (29) Hendriksen et al. (30) |
| Rabies | Antibody binding test ELISA |
Barth et al. (31) Adamovicz (32) |
| Yellow Fever | Plaque counting assy | WHO (33) |
The interest of the scientific community and financial support from public authorities are necessary for further development, validation, and introduction of alternatives. Animal experimentation in vaccine development, production, and quality control has contributed significantly to improvements in public health. This fact does not mean, however, that alternatives could not make an equal, or greater, contribution.
References
Bateson P: When to experiment on animals. New Scientist, 1986, Feb 20, p 30.
2. Orlans FB: Review of experimental protocols classifying animal harm and applying "refinements." Lab Anim Sci Jan 1987 (Special Issue), p 50.
3. Sharpe R: The cruel deception, in Fox MW, Mickley LD (eds): Advances in Animal Welfare Science 1986/1987. Boston, Martinus Nijhoff, 1987.
4. Miller EB: Comparative medicine: American experience with equine tetanus. Bull Hist Med 1983;57:81-92.
5. Niemi SM: Salmon and the first bacterin. Mod Vet Pract 1981;62:913-917.
6. Turner J: Reckoning with the Beast. Baltimore, Johns Hopkins Univ Pr, 1980.
7. Zo doende: (Animal Experimentation in the Netherlands, Statistics). Veterinary Public Health Inspectorate, Animal Experiments Department, Rijswijk, the Netherlands, 1989.
8. Henderson RH: The Expanded Programme on Immunization of the World Health Organization. Rev Infect Dis 1984;6(Suppl 2):S475-S479.
9. Stevenson LG 1955: Science down the drain. Bull Hist Med 1955;29:1-26.
10. Illich I: Medical Nemesis. New York, Pantheon Books, 1975.
11. McKeown T: The Role of Medicine. Oxford, Basil Blackwell, 1979.
12. McKinlay JB, McKinlay S: Health and Society. Millbank Memorial Fund, 1977.
13. Almquist E: Uber abnehmende Sterblichkeit und ihre veranlassende Ursachen (On decreasing mortality figures and their underlying causes). Zeitschrift für Hygiene 1888;4:1-19.
14. Banatvala JE: Measles must go and with it rubella. Brit Med J 1987;295:2-3.
15. Foege WH: Banishing measles from the world. World Health Forum 1984;5(1): 64.
16. Galazka AM, Lauer BA, Henderson RH, Keja J: Indications and contraindications for vaccines used in the Expanded Programme on Immunization. Bull World Health Organization 1984;62:357.
17. Cody CL, Baraff LJ, Cherry JD, Marcy SM, Manclark CR Nature and rates of adverse reactions associated with DTP and DT immunizations in infants and children. Pediatrics 1981;68:650-660.
18. Mortimer Jr EA: Pertussis immunization: Problems, perspectives, prospects. Hospital Practice. Oct, 1980, pp 103-118.
19. Koplan JP, Schoenbaum SC, Weinstein MC, Fraser DW: Pertussis vaccine - An analysis of benefits, risks and costs. N Engl J Med 1979;301:906-911.
20. Anon: Annual review of communicable diseases: 1982. Communicable Disease Report, Dec 1984, pp 1-2.
21. Stanfleld JP, Galazka A: Neonatal tetanus -- an underreported scourge. World Health Forum 1985;6:127-129.
22. Poore P: Vaccination strategies in developing countries. Vaccine 1988;6:393-398.
23. Abdussalam M: Animals for biomedical research: Perspective of developing countries, in XVIIth CIOMS Round Table Conference. Geneva, Dec 8-9, 1983.
24. Martin M: The price of polio vaccine. Lab Animal 1981;10(1):20-25.
25. Nicoll A: Contraindications to whooping-cough immunisation - Myths or realities. The Lancet, Mar 23, 1985, pp 679-681.
26. Hendriksen CFM: Laboratory Animals in Vaccine Production and Control Replacement, Reduction and Refinement. Boston, Kluwer Academic, 1988.
27. Kreeftenberg JG, van der Gun JW, Marsman FR, Sekhuis VM, Bhandari SK, Maheswari SC: An investigation of a mouse model to estimate the potency of the diphtheria component in combined vaccines. J Biol Stand 1985;13:229-234.
28. Hendriksen CFM, van der Gun JW, Kreeftenberg JG: The use of the Toxin Binding Inhibition (ToBI) test for the estimation of the potency of the diphtheria component of vaccines. J Biol Stand 1989;17:241-247.
29. Lyng J, Bentzon MW: Quantitative estimation of diphtheria and tetanus toxoids. 1. The flocculation test and the Lf-unit. J Biol Stand 1987;15:27-37.
30. Hendriksen CFM, van der Gun JW, Marsman FR, Kreeftenberg JG: The use of the in vitro Toxin Binding Inhibition (ToBI) Test for the estimation of the potency of tetanus toxoid. Biologicals 1991;19:23-29.
31. Barth R, Gross-Albenhausen E, Jaeger O, Milcke L: The antibody binding test, a useful method for quantitative determination of inactivated rabies virus antigen. J Biol Stand 1981;9:81-89.
32. Adamovicz P: The use of various immunochemical, biochemical and biological methods for the analysis of rabies virus production in tissue cultures. Develop Biol Standard 1984;55:191-197.
33. World Health Organization: Requirements for Yellow Fever Vaccine. WHO Technical Report Series 594.